[
https://issues.apache.org/jira/browse/HIVE-18398?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16447382#comment-16447382
]
Manoj Narayanan commented on HIVE-18398:
----------------------------------------
Looks like this behavior is different in Hive 2.X version due to fix done in
https://issues.apache.org/jira/browse/HIVE-12506. That Jira says that fix was
committed to branch 2.1, 2.0.0 and 1.3. I am not sure if 1.3 ever went out.
Anyhow As per
[https://cwiki.apache.org/confluence/display/Hive/HowToContribute#HowToContribute-UnderstandingHiveBranches]
and
[https://cwiki.apache.org/confluence/display/Hive/Home#Home-HiveVersionsandBranches]
what I understand is that branch-1 acts as source of all 1.X releases now.
branch-1 or branch-1.X doesn't contain that commit.
So I think commit for above Jira
[https://github.com/apache/hive/commit/ed9736080aa960ad7d7c2af58d692e6b4f9d6df5]
would need to be back ported to latest hive 1.X branch (branch-1)
> WITH SERDEPROPERTIES option is broken without
> org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe
> ------------------------------------------------------------------------------------------------
>
> Key: HIVE-18398
> URL: https://issues.apache.org/jira/browse/HIVE-18398
> Project: Hive
> Issue Type: Bug
> Components: Hive
> Affects Versions: 1.2.1
> Reporter: Rajkumar Singh
> Priority: Minor
>
> *Steps to reproduce:*
> 1. Create table
> {code}
> create table test_serde(id int,value string) ROW FORMAT DELIMITED FIELDS
> TERMINATED BY '|' ESCAPED BY '\\'
> {code}
> 2. show create table produce following output
> {code}
> CREATE TABLE `test_serde`(
> `id` int,
> `value` string)
> ROW FORMAT DELIMITED
> FIELDS TERMINATED BY '|'
> WITH SERDEPROPERTIES (
> 'escape.delim'='\\')
> STORED AS INPUTFORMAT
> 'org.apache.hadoop.mapred.TextInputFormat'
> OUTPUTFORMAT
> 'org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat'
> LOCATION
> 'hdfs://hdp262a.hdp.local:8020/apps/hive/warehouse/test_serde'
> TBLPROPERTIES (
> 'COLUMN_STATS_ACCURATE'='{\"BASIC_STATS\":\"true\"}',
> 'numFiles'='0',
> 'numRows'='0',
> 'rawDataSize'='0',
> 'totalSize'='0',
> 'transient_lastDdlTime'='1515448894')
> {code}
> 3. once you run the create table using the output of show create it ran into
> the parsing error
> {code}
> NoViableAltException(296@[1876:103: ( tableRowFormatMapKeysIdentifier )?])
> at org.antlr.runtime.DFA.noViableAlt(DFA.java:158)
> at org.antlr.runtime.DFA.predict(DFA.java:116)
> .....
> FAILED: ParseException line 6:0 cannot recognize input near 'WITH'
> 'SERDEPROPERTIES' '(' in serde properties specification
> {code}
> 4. table create with LazySimpleSerde don't have any such issue.
> {code}
> hive> CREATE TABLE `foo`(
> > `col` string)
> > ROW FORMAT SERDE
> > 'org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe'
> > WITH SERDEPROPERTIES (
> > 'serialization.encoding'='UTF-8')
> > STORED AS INPUTFORMAT
> > 'org.apache.hadoop.mapred.TextInputFormat'
> > OUTPUTFORMAT
> > 'org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat' ;
> OK
> Time taken: 0.375 seconds
> {code}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
